a-adrenergic Reduction of Cyclic Adenosine Monophosphate Concentrations in Rat Myocardium

Transcription

1 596 a-adrenergic Reduction of Cyclic Adenosine Monophosphate Concentrations in Rat Myocardium AUGUST M. WATANABE, DAVID R. HATHAWAY, HENRY R. BESCH, JR., BARBARA B. FARMER, AND ROBERT A. HARRIS SUMMARY We determined the effect of a-adrenergic receptor stimulation on cyclic adenosine monophosphate (cyclic AMP) concentrations in isolated myocytes derived from adult rat hearts and in isolated perfused rat hearts. Activation of a-adrenergic receptors with either phenylephrine (10~ 8 M to 10~ 6 M) or epinephrine (10" M to 10~ 6 M) plus propranolol (10 '! M) resulted in a reduction in cyclic AMP levels in isolated myocytes. The action of phenylephrine was antagonized by phentolamine (10~ 6 M). Phenylephrine (10 s M) attenuated cyclic AMP generation in response to isoproterenol (10" s M and 10 s M). However, this effect of phenylephrine was not antagonized by phentolamine. Elevation of cyclic AMP concentrations produced by glucagon and by theophylline in isolated myocytes was attenuated by phenylephrine and by epinephrine plus propranolol and the attenuation was antagonized by phentolamine. In isolated perfused rat hearts epinephrine (10 (i M), when given with propranolol, diminished the rate of development of tension and also reduced tissue levels of cyclic AMP. Epinephrine alone, as well as isoproterenol, increased contractility and myocardial cyclic AMP concentrations as expected. These results indicate that catecholamines may increase or decrease cyclic AMP levels in rat myocardium, depending on the intensity of stimulation of receptor types. Increases are mediated by /3-adrenergic receptors, whereas decreases appear to be mediated by a-adrenergic receptors. IN 1967, Robison et al. 1 hypothesized that activation of either a-adrenergic or /3-adrenergic receptors may modify the activity of adenylate cyclase and result in changes in the rate of conversion of ATP to cyclic adenosine monophosphate (cyclic AMP). They further predicted that a- sympathomimetic agents could induce a decrease in the activity of the enzyme in some tissues. This original hypothesis has since been corroborated by studies which indicate that some a-adrenergic effects are mediated by a reduction in the tissue level of cyclic AMP. 2 " 5 Although there is an abundance of evidence that activation of a-adrenergic receptors can lead to alterations in the electrophysiological and mechanical functions of the heart, 6 " 18 no intracellular mediator or biochemical basis for these effects has been identified. It has been fairly convincingly shown that these a-adrenergic effects are not mediated by an increase in myocardial cyclic AMP concentrations The question remains, however, whether any of the a-adrenergic effects on the heart might be mediated by a decrease in tissue cyclic AMP levels. The purpose of the present study was to attempt to demonstrate whether stimulation of myocardial a-adrenergic receptors is associated with a fall in cyclic AMP concentrations. The studies were carried out using isolated adult rat heart cells. To extend the findings to the intact heart, we also performed a limited parallel study of a-adrenergic From the Departments of Medicine, Pharmacology, and Biochemistry, and The Krannert Institute of Cardiology, Indiana University School of Medicine, Indianapolis, Indiana. Supported by U. S. Public Health Service Grants HL 06308, HL 05363, HL 18795, HL 07182; Specialized Center of Research in Hypertension (SCOR) HL 14159; The American Heart Association, Indiana Affiliate; The Herman C. Krannert Fund; and The Showalter Residuary Trust. Address for reprints: August M. Watanabe, M.D., Indiana University School of Medicine, 1100 West Michigan Street, Indianapolis, Indiana Received April 7, 1976; accepted for publication December 9,1976. effects on contractility and cyclic AMP levels of isolated, perfused rat hearts. ISOLATION OF CELLS Methods Our method 21 for isolating cells is a modification of that described by Berry et al. 22 Briefly, this involves Langendorff perfusion of rat hearts at 37 C with a well oxygenated, nominally Ca 2+ -free and Mg 2+ -free buffer solution containing (ITIM): NaCl, 134; KC1, 5; sodium phosphate (ph 7.4), 6; and glucose, 10. After about 5 minutes, collagenase (80 U/ml) and hyaluronidase (82 U/ml) were added to the perfusion fluid and the perfusion was continued at 37 C for an additional 20 minutes. The hearts then were removed from the perfusion apparatus, sliced, and incubated in a shaking bath for a further 10 minutes in fresh perfusion fluid. The disaggregated tissue was quickly cooled and then filtered through nylon mesh. The cells were collected from the filtrate by centifugation at a force of 37 g, washed twice by resuspension and recentrifugation in an incubation medium that contained (ITIM): NaCl, 120; KCl,5;MgCl 2, 1.2; sodium phosphate (ph 7.4), 10; and glucose, 5. Further details of the method are described by Farmer et al. 21 Cell viability was determined by the criterion of trypan blue exclusion: 23 the percentage of cells in three high power fields which were able to exclude 0.3% trypan blue was determined under a microscope using a hemocytometer counting chamber and a hand tally counter. EXPERIMENTS WITH ISOLATED CELLS The concentration of cells harvested by the isolation procedure was determined by quantifying the protein content per unit volume of suspended cells. For experiments, the isolated cells were resuspended to a final concentration of 4-6 mg dry weight of cells per ml. The cells (1-ml

2 a-adrenergic REDUCTION OF CYCLIC AMP IN HEART/Watanabe el al. 597 samples) then were pipetted into 25-ml siliconized Erlenmeyer flasks containing 2 ml of phosphate HEPES buffer which had been oxygenated with 100% oxygen for 60 minutes prior to use. The flasks then were placed in a Dubnoff shaking water bath heated to 37 C. All flasks were allowed to preincubate in the water bath for 5 minutes to bring the temperature of the cells to 37 C. After the 5-minute preincubation period drugs were added for the actual experiment. PERFUSION OF WHOLE HEARTS Hearts from rats of either sex weighing g were perfused by the Langendorff method as previously described. The modified Krebs-Henseleit perfusion buffer contained (HIM): NaCl, 118; KC1, 4.8; MgSO 4, 1.2; KH 2 PO 4, 1.0; NaHCO 3, 27.2; CaCl 2, 2.0 or 2.5; and glucose, The temperature was maintained at 37 C. Hearts were paced with an electronic stimulator at a rate of 210 stimuli/min. Diastolic tension was set at 5 g before beginning the experiment. Developed tension, its first derivative (dt/dt), and perfusion pressure were recorded continually throughout the experiment. The contractile responses were similar whether they were assessed from developed tension or dt/dt. All reported analyses of changes in contractility were made on the basis of changes in dt/dt. The hearts were perfused for at least 30 minutes before starting any experimental intervention. Drugs were administered through a side arm that entered the main perfusion stream about 2 cm above the coronary arteries of the heart. This arrangement allowed mixing and diluting of the drugs immediately prior to use and also did not necessitate prolonged oxygenation of the drugs. All of the drugs were administered by continuous infusions. ASSAY OF CYCLIC AMP Hearts in which cyclic nucleotide levels were to be measured were frozen quickly with metal tongs that previously had been cooled to the temperature of liquid nitrogen ( 196 C). Freeze-clamped hearts were pulverized and the powdered tissue was prepared for assay by homogenization in 5% trichloroacetic acid (TCA), which was then removed by extraction with ether. The experiments with isolated cells were terminated by the addition of 0.5 ml of 35 % TCA to the cell suspension (final concentration of TCA = 5%) followed by homogenization of the cells. The TCA was removed by extraction with ether. The extracted homogenates of both the whole hearts and isolated cells were then introduced into an ion-exchange column (Dowex AG 1 x 8, mesh, formate form) to remove substances that might interfere with the assay of cyclic AMP. Cyclic AMP concentrations were measured by a competitive protein-binding assay 24 using binding protein derived from beef adrenal glands, 25 as previously described. 26 CHEMICALS L-Isoproterenol bitartrate, L-epinephrine bitartrate, L- norepinephrine HO, L-phenylephrine HC1, cyclic AMP, theophylline, and hyaluronidase (purified from bovine testes) were obtained from Sigma; purified collagenase (derived from Clostridium histolyticum) from Worthington Biochemicals; DL-propranolol HC1 from Ayerst; crystalline glucagon from Lilly; phentolamine HC1 from Ciba; 3 H-cyclic AMP (specific activity, 37.7 Ci/mmol) from New England Nuclear. STATISTICAL ANALYSIS The statistical significance of differences between means was determined by unpaired /-tests. 27 Results VIABILITY OF ISOLATED CELLS A detailed description of the morphology and biochemical properties of isolated rat heart cells is presented elsewhere. 21 Only those features of the cells which are directly relevant to this study are considered in this paper. Assessed within 5 minutes after preparing the cells, the percent of the total number of cells which were viable, as judged by their ability to exclude 0.3% trypan blue, was 72 ± 1 % (mean ± SEM) for eight separate preparations. At Ca 2+ concentrations of 10" 5 M or less, the cells remained viable for at least 30 minutes when incubated at 37 C. The isolated cells were sensitive to the Ca 2+ concentration in the incubation medium. When the Ca 2+ concentration was increased to 10~ 4 M, viability of the cells decreased to about 45% in 5 minutes and about 30% in 12 minutes. Therefore, all cells were incubated in a medium containing about 10~ 6 M Ca 2+. EFFECT OF CATECHOLAMINES ON CYCLIC AMP CONCENTRATIONS IN ISOLATED CELLS Cyclic AMP levels rapidly increase when cells are exposed to 10~ 7 M isoproterenol (Fig. 1, inset). Furthermore, the steady state level of cyclic AMP appears to be stable, during continuous exposure to isoproterenol, for at least 5 minutes. We obtained similar results with kinetic studies on the effect of norepinephrine (10~ 7 M) on cyclic AMP levels. Thus, most of the experiments with catecholamines were terminated after 2 minutes of incubation of cell with drugs. The isolated cells responded to catecholamines in the expected manner (Fig. 1). Isoproterenol was the most potent catecholamine that increased cyclic AMP levels: the threshold concentration for stimulation was between 10~ 9 M and 10~ 8 M, and 10~ 5 M produced a 250% increase in cyclic AMP concentrations. To further support the /3- adrenergic receptor mediation of increased cyclic AMP levels, cells were incubated with varying concentrations of isoproterenol simultaneously with 10~ 6 M propranolol (Fig. 2). Propranolol attenuated the response of the cells to isoproterenol in a competitive manner. EFFECT OF a-adrenergic RECEPTOR AGONISTS ON CYCLIC AMP CONCENTRATIONS IN ISOLATED CELLS As in our studies on /3-adrenergic receptor agonists, we performed preliminary experiments to determine the time course of the effects of a-adrenergic receptor agonists on cyclic AMP levels (Fig. 3, inset). Phenylephrine (10~ 7 M)

3 598 CIRCULATION RESEARCH VOL. 40. No. 6, June 1977 O OISOPROTERENOL A- AEPINEPHRINE D O NOREPINEPHRINE Duration of Phenylephrine Exposure (min) s U 4 3 *-l O -^3 PHENYLEPHRINE --# PHENYLEPHRINE PHENTOLAMINE 10" 6 M Duration of Isoprot. Exposure (min),o'-e 10-" CATECHOLAMINE CONCENTRATION (molar) FIGURE 1 Effect of catecholamines on cyclic AMP concentrations in isolated heart cells. Values represent means ± SEM of 4 to 16 preparations of cells for each point. Basal value = 3.9 ± 0.01 pmol/mg dry weight; n = 45. Inset: time course of the generation of cyclic AMP by heart cells during exposure to isoproterenol. Points are means of duplicates. appeared to reduce cellular cyclic AMP levels, but this effect developed much more slowly than the /3-adrenergic receptor-mediated stimulation of cyclic AMP levels. In the presence of phenylephrine, cyclic AMP levels remained unchanged for 1 minute. After 90 seconds of continuous exposure to phenylephrine, cyclic AMP levels began to fall and reached a minimum at 2 minutes, after which the steady state levels appeared to stabilize. The effect of phenylephrine, after 2 minutes of incubation, on cyclic AMP levels in isolated cells is shown in Fig- ISOPROTERENOL CONCENTRATION (molar) FIGURE 2 Effect of propranolol (10~ 6 M) on isoproterenol stimulation of cyclic AMP levels in heart cells. Each point is the mean ± SEM of 3 to 13 different preparations. Basal value = 3.9 ± 0.01 pmol/mg dry weight; n = 45. PHENYLEPHRINE CONCENTRATION (molar) FIGURE 3 Effect of phenylephrine alone and with phentolamine (10" 6 M) on cyclic AMP concentrations in isolated heart cells. Each point is the mean ± SEM of 4 to 7 preparations. Basal value =3.9 ± 0.01 pmol/mg dry weight; n =45. Inset: time course of the effect of phenylephrine on cyclic AMP levels in isolated cells. Each point is the mean of duplicates, a: P < compared to basal, b: P < compared to phenylephrine plus phentolamine. c: P < 0.05 compared to basal, d: P < 0.01 compared to basal. ure 3. At 10" 8 M to 10" 6 M concentrations of phenylephrine there was a small (21% at 10~ 6 M) reduction in cellular cyclic AMP levels. With 10~ 5 M and 10" 4 M phenylephrine there was a small increase in cyclic AMP concentration. The competitive a-adrenergic receptor blocker phentolamine (10~ 6 M) abolished the reduction in cyclic AMP concentration caused by phenylephrine but did not significantly attenuate its ability to elevate the cyclic AMP levels. Phentolamine alone (10"' M to 10~ 5 M) did not affect cyclic AMP concentrations. Epinephrine is a mixed-adrenergic receptor agonist. Of the naturally occurring catecholamines, it is the most potent agonist of both /3- and a-adrenergic receptors. 28 However, if its effect on one receptor type predominates, effects of its interaction with other receptors may be obscured. In an attempt to unmask obscured effects we performed a series of experiments using epinephrine plus propranolol (10~ 6 M) to see whether we could elicit an a- adrenergic effect of the catecholamine. When /3-adrenergic receptors were blocked, epinephrine reduced cyclic AMP levels, an effect opposite to that found when the catecholamine was given alone (Fig. 4). Between the concentrations of 10" 8 M and 10" 6 M, epinephrine given in the presence of propranolol reduced cyclic AMP levels, the maximal effect occurring at 10" 6 M. At this concentration, epinephrine reduced the nucleotide level 28% below basal values. Propranolol alone (10~ 7 M to 10" 5 M) had no effect on cellular cyclic AMP levels. EFFECT OF PHENYLEPHRINE ON ISOPROTERENOL- INDUCED INCREASES IN CYCLIC AMP CONCENTRATIONS OF ISOLATED CELLS In other tissues it has been found that the reduction of cyclic AMP levels induced by a-adrenergic agents is more pronounced when cyclic AMP generation has been simultaneously stimulated. Consequently, we performed experiments to determine the effect of activating a-adrenergic

4 a-adrenergic REDUCTION OF CYCLIC AMP IN HEART/Watanabe el al. 599 PHENTOLAMINE 10" 6 M o 3 a. S < " '' ' " 'I-* O 2-45.v.v..:10V PHENYLEPHRINE EPINEPHRINE CONCENTRATION (molar) FIGURE 4 Effect ofepinephrine atone and with propranolol (10'* M) on cyclic AMP levels in isolated heart cells. Each point is the mean ± SEMof4 to 16 different preparations. Basal value = 3.9 ± 0.01 pmol/mg dry weight: n = 45. a: P <0.0I compared to basal, b: P < compared to basal. GLUCAGON 10" 7 M FIGURE 6 Effect of glucagon alone and with phenylephrine on cyclic AMP levels in heart cells. Values are mean ± SEM of the number of preparations noted at the bottom of the bars, a: P < 0.05 compared to glucagon alone, b: P < 0.05 compared to phenylephrine alone. receptors during isoproterenol-induced stimulation of cyclic AMP levels. Phenylephrine (10~ 7 M) was without effect on the increase in cyclic AMP concentrations induced by isoproterenol (Fig. 5). However, 10~ 5 M phenylephrine attenuated the generation of cyclic AMP in response to two concentrations of isoproterenol, 10" 8 M and 10" 5 M. The a-adrenergic receptor antagonist phentolamine did not significantly alter this effect of phenylephrine. EFFECT OF a-adrenergic AGONISTS ON NONADRENERGIC AUGMENTATION OF CYCLIC AMP CONCENTRATIONS IN ISOLATED CELLS The experiments described in the previous section were suggestive but did not prove that a-adrenergic agonists ISOPROTERENOL 10" M PHENYLEPHRINE ISOPROTERENOL 10" 5 M FIGURE 5 Effect of phenylephrine on isoproterenol stimulation of cyclic AMP levels in heart cells. Values are mean ± SEM. The number of preparations studied is signified by the numbers at the bottom of the bars, a: P < 0.01 compared to isoproterenol alone, b: P < compared to isoproterenol alone. could attenuate cyclic AMP generation induced by isoproterenol: phentolamine did not reverse the effect of phenylephrine. Propranolol could not be used in these experiments, therefore we were limited to the use of phenylephrine as the a-adrenergic receptor agonist. In the series of experiments described in this section we augmented cyclic AMP levels independently of /3-adrenergic receptor stimulation, by activating adenylate cyclase with glucagon and by inhibiting the breakdown of cyclic AMP with the phosphodiesterase inhibitor theophylline. Phenylephrine in two concentrations (10~ 7 M and 10~ 6 M) significantly attenuated cyclic AMP generation in response to glucagon (Fig. 6). Phentolamine (10~ 6 M) abolished this effect of 10" 7 M phenylephrine. Similar results were obtained when cyclic AMP levels were increased by administration of theophylline. Both phenylephrine (10~ 7 M) and epinephrine (10~ 7 M) plus propranolol (10~ 6 M) attenuated the theophylline-induced stimulation of cyclic AMP levels (Fig. 7). Phentolamine (10~ 6 M) tended to reverse this effect of the a-adrenergic agonists. EFFECT OF a-adrenergic RECEPTOR BLOCKADE ON EPINEPHRINE-INDUCED AUGMENTATION OF CYCLIC AMP CONCENTRATIONS IN ISOLATED CELLS If activation of a-adrenergic receptors reduces cyclic AMP levels or attenuates the generation of the nucleotide in response to stimulating agents, it is possible that the mixed a- and /3-receptor agonist epinephrine is inhibiting its own ability to stimulate cyclic AMP levels. To test this possibility, we performed experiments to determine whether blockade of a-adrenergic receptors would potentiate epinephrine-induced elevation of cellular cyclic AMP levels. There appeared to be a slight increase in the response of cells of epinephrine in the presence of phentolamine. However, at no point was the change statistically significant.

5 600 CIRCULATION RESEARCH VOL. 40. No. 6. June 1977 A & EPINEPHRINE 10" 6 M -" EPINEPHRINE 10" 6 M + PROPRANOLOL 3 x 10" 6 M- " PHENYL. 10* 7 M EPI. 10' 7 M + PROP. 10" 6 M > ^ / THEOPHYLLINE 10" 3 M FIGURE 7 Effect of theophylline alone and with the a-adrenergic agonists phenylephrine and epinephrine plus propranolol on cyclic AMP levels in isolated myocytes. Values are mean ± SEM of the number of preparations noted at the bottom of the bars, a: P < 0.05 compared to theophylline alone. EFFECT OF o-adrenergic RECEPTOR STIMULATION ON MECHANICAL FUNCTION AND ON CYCLIC AMP LEVELS IN WHOLE HEARTS Experiments were performed on isolated perfused rat hearts at several different Ca 2+ concentrations. A negative inotropic effect which might be expected to accompany a reduction in cyclic AMP levels could be seen only when hearts were perfused with 2.0 or 2.5 mm Ca 2+. At lower, more physiological Ca 2+ concentrations, no negative inotropy was observed. When the hearts were perfused with the high Ca 2+ concentrations, 10^6 M epinephrine in the presence of propranolol (3 x 10~ 6 M) produced a slowly developing negative inotropic effect which became maximal after about 7 minutes of perfusion with this drug combination (Fig. 8). In these experiments, propranolol alone sometimes produced a small negative inotropic effect. By minutes of its infusion, however, contractility stabilized and then the epinephrine infusion was begun. Epinephrine (10~ 6 M) alone always produced the expected increase in contractility (Fig. 8). In a series of hearts preperfused with 10~ 5 M propranolol, 10~ 6 M epinephrine reduced tissue cyclic AMP levels below basal values (Fig. 9). Epinephrine alone and isoproterenol increased tissue cyclic AMP concentrations, as expected. Propranolol, by itself, did not alter the levels. Discussion During the past decade it has been demonstrated that catecholamines can reduce cyclic AMP levels in several different tissues derived from a variety of species. 2 " 5 29> 30 ' This effect appeared in each case to be mediated by the a- adrenergic receptor because it was elicited by a-receptor agonists (norepinephrine or epinephrine) and/or blocked by a-antagonists (phentolamine, phenoxybenzamine, dihydroergotamine, or ergotamine). Furthermore, in several of these studies the functional response associated with the fall in cyclic AMP level was opposite that associ- DURATION OF EPINEPHRINE PERFUSION (min) FIGURE 8 Effect of epinephrine (10 M) alone and with propranolol (3 x 10~" M) on dt/dt of isolated perfused rat hearts. Each point represents mean ± SEM of eight hearts for epinephrine alone and seven hearts for epinephrine plus propranolol. ated with a rise in cyclic AMP concentration From these data, it appears that in some tissues catecholamines can either increase or decrease cyclic AMP concentrations, depending on whether they interact with /3- or a- adrenergic receptors, and these changes in nucleotide levels may be associated with reciprocal functional effects. It is generally accepted that cyclic AMP mediates the metabolic effects of catecholamines on the heart, 31 and a large body of data suggests that this second messenger mediates the inotropic 26 and chronotropic effects of catecholamines as well, although some controversy persists on this latter point. 34 The present investigation was specifically designed to attempt to answer whether catecholamines can decrease, as well as increase, cyclic AMP levels in the myocardium. Our data indicate that a-adre- BASAL PJOP! 10-"M E P>- 10 " 6M "O. 10"'M Prop. 10" 6 M Prop. 10" 5 M FIGURE 9 Effect of catecholamines and propranolol on cyclic AMP levels in isolated perfused rat hearts. Values are mean ± SEM of the number of hearts noted at the bottom of the bars, a: P <0.01 compared to basal, b: P < compared to basal.

6 a-adrenergic REDUCTION OF CYCLIC AMP IN HEART/Watanabe et al. 601 nergic receptor stimulation can indeed result in a reduction in cyclic AMP levels. When a-receptors were stimulated in presumably basal or "resting" cardiac cells, the fall in cyclic AMP levels was very small. However, when the cyclic AMP levels in the cells were increased, the a- adrenergic reduction of cyclic AMP was magnified, an interaction that might be described as accentuated antagonism. Similar results have been found in other tissues. In fact, in some tissues, e.g., human platelets, and a- adrenergic receptor-mediated reduction in cyclic AMP level becomes apparent only after the concentration of the nucleotide has been increased by another agent. 4 a-adrenergic receptor stimulation attenuated the generation of cyclic AMP in response to glucagon, which interacts with putative glucagon receptors coupled to adenylate cyclase. It also reduced cyclic AMP elevation induced by theophylline, an agent which is thought to act by inhibiting phosphodiesterase. Such a generalized effect is compatible with a-adrenergic receptor inhibition of adenylate cyclase. If the mechanism by which a-adrenergic receptor stimulation reduced cyclic AMP levels were an increase in phosphodiesterase activity, it seems unlikely that the effect would persist in the face of maximal inhibition of phosphodiesterase by a high concentration of theophylline. Our data from cardiac tissue support the conclusions of other investigators using other tissues that a- adrenergic agonists may act by inhibiting the enzyme adenylate cyclase.- * 35 The mechanism by which phenylephrine attenuated the effect of isoproterenol is unclear. If phenylephrine acted in this instance by stimulating a-adrenergic receptors, its effect should have been abolished by phentolamine. Phenylephrine is a weak /3-adrenergic receptor agonist, as described by other investigators 20 and confirmed by our own studies, and thus it could have acted as a /3-adrenergic receptor antagonist when given with isoproterenol. We cannot explain the different action of phenylephrine when given with isoproterenol than when given with glucagon or theophylline. It has been suggested that the /3- and a- adrenergic receptors are very closely related and may even represent two allosteric configurations of a single active site. 36 In line with this notion, results of some studies have been interpreted as showing transformation of f3- to a- adrenergic receptors. 36 Perhaps, in our study, when /3- adrenergic receptors were stimulated by isoproterenol, a- adrenergic receptors became inaccessible to phenylephrine. Such a phenomenon could also explain the lack of potentiation of epinephrine augmentation of cyclic AMP levels by phentolamine. If the mixed-adrenergic agonist epinephrine was limiting its own ability to stimulate adenylate cyclase, a-adrenergic receptor blockade should have enhanced the epinephrine effect, as has been observed in several other studies. However, such was not the case in our study. A number of investigators have demonstrated that a- adrenergic receptor stimulation produces both electrophysiological and inotropic effects on the heart. A principle electrophysiological effect is a prolongation of the total duration of the monophasic action potential and is functionally manifested as a prolongation of the refractory period. 9 - " With regard to the inotropic effects, the majority of the studies show that a-adrenergic receptor stimulation increases myocardial contractility. l0> ~ 18 The mechanism by which a-adrenergic receptor stimulation increases contractility appears to be quite different from the mechanism for the /3-adrenergic receptor stimulation. During a-adrenergic receptor stimulation, cyclic AMP levels are not increased, 19 ' 20 and many of the mechanical effects associated with a-adrenergic stimulation are different from those of /3-adrenergic receptor stimulation. In additon to the well known positive inotropic effects, several investigators have demonstrated a cardiodepressant effect of a-adrenergic receptor stimulation. Imai et al. 7 in 1961 and James and colleagues 13 in 1968 showed that the a-adrenergic receptor agonist methoxamine slowed heart rate and antagonized the cardioaccelerating effect of exogenously administered or endogenously released catecholamines. The former group also showed that methoxamine produced a negative inotropic effect on dog hearts and abolished the positive inotropic and chronotropic actions of epinephrine. 6 A negative inotropic effect of a-adrenergic receptor stimulation has since been observed by several other investigators " Thus, a-adrenergic receptor stimulation may result in either an increase or a decrease in the inotropic state of the heart. These observations prompted Wenzel and Su 8 to postulate that there are a- and ^-stimulatory receptors and a-inhibitory receptors in the myocardium. It appears that the negative inotropic effect of a-inhibitory receptors become manifest only under certain conditions, perhaps associated with a high cellular Ca 2+ level. For example, Endoh and Schumann 17 showed that methoxamine produced a negative inotropic effect in rabbit papillary muscle only during relatively rapid (1.5 Hz) stimulation, and we found that epinephrine plus propranolol produced a negative inotropic effect only if the hearts were perfused with buffer containing 2.0 or 2.5 mm Ca 2+, a level about twice the physiological concentration in rats. The physiological significance of our findings in isolated cells remains to be established. In the isolated cells it was not possible to determine the functional significance of the observed reduction in nucleotide levels. Our correlative study in whole hearts was performed to attempt to answer the question of whether a-adrenergic reduction of cyclic AMP levels occurs in intact cardiac tissue and to determine what functional significance lowering cyclic AMP levels might have. We found that when activation of a- adrenergic receptors depressed contractility, there was an associated reduction in tissue cyclic AMP levels in the hearts. As previously mentioned, several other investigators have also demonstrated a negative inotropic and chronotropic effect of a-adrenergic receptor stimulation on whole hearts. 6 " ' 17 Recently, Posner and co-workers 37 and Danilo and colleagues 38 showed that activation of a-adrenergic receptors decreased spontaneous rates of isolated dog Purkinje fibers. Taken together, these diverse studies indicate that activation of a-adrenergic receptors can, under certain conditions, lead to a functional depression of cardiac tissue. If an increase in cyclic AMP levels

7 602 CIRCULATION RESEARCH VOL. 40, No. 6, JUNE 1977 does mediate the cardiac stimulatory effects of catecholamines, a reduction in cyclic AMP levels could logically be expected to produce the opposite effect. It thus appears possible that a-inhibitory receptors produce their cardiodepressant effects by reducing tissue cyclic AMP levels. Acknowledgments We gratefully acknowledge the expert technical assistance of Bruce Henry and Carolyn Hicks and the secretarial assistance of Carol Holliday. We thank the following for their gifts of drugs used: Ayerst Laboratories for DL-propranolol, Eli Lilly and Company for crystalline glucagon, and Ciba Pharmaceutical Company for phentolamine HO. References 1. Robison GA, Butcher RW, Sutherland EW: Adenyl cyclase as an adrenergic receptor. Ann NY Acad Sci 139: , Handler JS, BensingerR, Orloff J: Effect of adrenergic agents on toad bladder response to ADH, 3',5'-AMP, and theophylline. Am J Physiol 215: , Abe K, Robison GA, Liddle GW, Butcher RW, Nicholson WE, Baird CE: Role of cyclic AMP in mediating the effects of MSH, norepinephrine, and melatonin on frog skin color. Endocrinology 85: , Robison GA, Arnold A, Hartman RC: Divergent effects of epinephrine and prostaglandin E, on the level of cyclic AMP in human blood platelets. Pharmacol Res Commun 1: , Austen KF: Reaction mechanisms in the release of mediators of immediate hypersensitivity from human lung tissue. Fed Proc 33: , Imai S, Shigei T, Hashimoto K: Cardiac actions of methoxamine with special reference to its antagonistic action to epinephrine. Circ Res 9: , Imai S, Shigei T, Hashimoto K: Antiaccelerator action of methoxamine. Nature 189: , Wenzel DG, Su JL: Interactions between sympathomimetic amines and blocking agents on the rat ventricle strip. Arch Int Pharmacodyn Ther 160: , Govier WC, Mosal NC, Whittington P, Broom AH: Myocardial alpha and beta adrenergic receptors as demonstrated by atrial functional refractory-period changes J Pharmacol Exp Ther 154: , Govier WC: A positive inotropic effect of phenylephrine mediated through alpha adrenergic receptors. Life Sci 6: , Benfey BG, Varma DR: Interactions of sympathomimetic drugs, propranolol and phentolamine, on atrial refractory period and contractility. J Pharmacol Chemother 30: , Govier WC: Myocardial alpha adrenergic receptors and their role in the production of a positive inotropic effect by sympathomimetic agents. J Pharmacol Exp Ther 159: 82-90, James TN, Bear ES, Lang RF, Green EW: Evidence for adrenergic alpha receptor depressant activity in the heart. Am J Physiol 215: , Pappano AJ: Propranolol-insensitive effects of epinephrine on action potential repolarization in electrically driven atria of the guinea pig. J Pharmacol Exp Ther 177: 85-95, Benfey BG: Characterization of a-adrenoceptors in the myocardium. Br J Pharmacol 48: , Schumann HJ, Endoh M, Wagner J: Positive inotropic effects of phenylephrine in the isolated rabbit papillary muscle mediated both by a- and 0-adrenoceptors. Naunyn Schmiedebergs Arch Pharmacol 284: , Endoh M, Schumann HJ: Frequency-dependence of the positive inotropic effect of methoxamine and naphazoline mediated by a-adrenoceptors in the isolated rabbit papillary muscle. Naunyn Schmiedebergs Arch Pharmacol 287: , Ledda F, Marchetti P, Mugelli A: Studies on the positive inotropic effect of phenylephrine; a comparison with isoprenaline. Br J Pharmacol 54: 83-90, Benfey BG, CarolinT: Effect of phenylephrine on cardiac contractility and adenyl cyclase activity. Can J Physiol Pharmacol 49: , McNeil] JH, Verma SC: Phenylephrine-induced increases in cardiac contractility, cyclic adenosine monophosphate and phosphorylase a. J. Pharmacol Exp Ther 187: , Farmer BB, Harris RA, Jolly WW, Hathaway DR, Katzberg A, Watanabe AM, Whitlow AL, Besch JR Jr: Isolation and characterization of adult rat heart cells. Arch Biochem Biophys 1976 (in press) 22. Berry MN, Friend DS, Scheuer J: Morphology and characterization of intact muscle cells isolated from adult rat heart. Circ Res 26: , Howard RB, Pesch LA: Respiratory activity of intact, isolated parenchymal cells from rat liver. J Biol Chem 243: , Gilman AG: Protein binding assay for adenosine 3',5'-cyclic monophosphate. Proc Natl Acad Sci USA 67: , Brown BL, Albano JDM, Ekins RP, Sgherzi AM: Simple and sensitive assay method for the measurement of adenosine 3':5' cyclic monophosphate. Biochem J 121: , Watanabe AM. Besch HR Jr: Cyclic adenosine monophosphate modulation of slow calcium influx channels in guinea pig hearts. Circ Res 35: , Dixon WF, Massey FJ Jr: Introduction to Statistical Analysis, ed 3. New York, McGraw-Hill, 1969, pp Goodman LS, Gilman A: The Pharmacological Basis of Therapeutics, ed 5. New York, Macmillan, 1975, pp Robison GA, Langley PE, Burns TW: Adrenergic receptors in human adipocytes; divergent effects on adenosine 3',5'-monophosphate and lipolysis. Biochem Pharmacol 21: , Turtle JR, Kipnis DM: An adrenergic receptor mechanism for the control of cyclic 3',5'-adenosine monophosphate synthesis in tissues. Biochem Biophys Res Commun 28: 797, Sutherland EW, Robison GA, Butcher RW: Some aspects of the biological role of adenosine 3',5'-monophosphate (cyclic AMP). 1 Circulation 37: , Tsien RW, Giles W, Greengard P: Cyclic AMP mediates the effects of adrenaline on cardiac Purkinje fibers. Nature (New Biol) 240: , Danilo P, Vulliemoz Y, Verosky M, Rosen R, Hoffman BF: Epinephrine-induced automaticity and camp concentration in canine cardiac Purkinje fiber bundles (abstr). Fed Proc 33: 479, Sobel BE, Mayer SE: Cyclic adenosine monophosphate and cardiac contractility. Circ Res 32: , Burns TW, Langley PE: The effect of alpha and beta adrenergic receptor stimulation on the adenylate cyclase activity of human adipocytes. J Cyclic Nucleotide Res 1: , Kunos G, Yong MS, Nickerson M: Transformation of adrenergic receptors in the myocardium. Nature (New Biol) 241: , Posner P, Lambert CR, Farrar EL: Biphasic effects of epinephrine on potassium transport and automaticity in cardiac Purkinje fibers (abstr). Circulation 50 (suppl III): 146, Danilo P, Hordof AJ, Ilvento JP, Alonso MB, Rosen MR: Age related changes in automaticity and catecholamine responsiveness of canine cardiac Purkinje fibers. Fed Proc 35: 22, 1976